Cardiovascular Surgery

Cardiovascular Surgery
Effect of Timing of Chronic Preoperative Aspirin
Discontinuation on Morbidity and Mortality in Coronary
Artery Bypass Surgery
Miriam Jacob, MD; Nicholas Smedira, MD; Eugene Blackstone, MD;
Sarah Williams, MS; Leslie Cho, MD
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
Background—Aspirin (ASA) has been shown to reduce postoperative coronary artery bypass grafting (CABG) mortality
and ischemic events; however, the timing of chronic ASA discontinuation before surgery is controversial because of
concern about postoperative bleeding. We evaluated the effect of the timing of ASA discontinuation before CABG on
major adverse cardiovascular outcomes and postoperative bleeding using the Cleveland Clinic Cardiovascular
Information Registry database.
Methods and Results—At the Cleveland Clinic between January 1, 2002, and January 31, 2008, 4143 patients undergoing
CABG were taking preoperative chronic ASA. Of these, 2298 discontinued ASA 6 or more days before surgery (early
discontinuation), and 1845 took ASA within 5 days of the surgery (late use). Because of substantial differences between
these 2 groups, propensity score analysis, and matching based on 31 variables were used for fair comparison of
outcomes. This resulted in 1519 well-matched pairs of patients (73%). There was no significant difference between those
with early discontinuation and late ASA use with regard to the composite outcome of in-hospital mortality, myocardial
infarction, and stroke (1.7% versus 1.8%, P⫽0.80). Late use was associated with more intraoperative transfusions (23%
versus 20%, P⫽0.03) and postoperative transfusions (30% versus 26%, P⫽0.009) but a similar number of reoperations
for bleeding (3.4% versus 2.4% P⫽0.10).
Conclusions—Among patients undergoing isolated CABG, late discontinuation of ASA resulted in no difference in
postoperative cardiovascular outcomes; however, there was an increased transfusion requirement. Thus, we recommend
weighing the risks and benefits of late ASA use in these patients. (Circulation. 2011;123:577-583.)
Key Words: aspirin 䡲 arteriosclerosis 䡲 bypass 䡲 coronary disease 䡲 surgery
A
spirin (acetylsalicylic acid [ASA]) is an integral chronic
therapy for patients with coronary artery disease. It has
been shown to reduce postoperative coronary artery bypass
graft (CABG) surgery mortality and ischemic events. However, there has been controversy regarding the timing of ASA
discontinuation before CABG surgery owing to concern
about postoperative bleeding complications. The use of early
postoperative aspirin is clear given its beneficial effect on
vein graft patency and the decrease in adverse postoperative
outcomes, including mortality, myocardial infarction (MI),
stroke, renal failure, and bowel infarction.1,2 Prior retrospective studies of isolated CABG suggest that use of ASA in the
preoperative period results in a significant reduction of
in-hospital mortality ranging from 45% to 66% with use
within 7 and 5 days of surgery, respectively.2,3
Editorial see p 571
Clinical Perspective on p 583
The recommendations regarding timing of ASA discontinuation differ between the cardiology and cardiovascular surgery
societies. The American College of Cardiology and American
Heart Association 2004 guideline update for CABG states that in
stable patients, ASA should be discontinued 7 to 10 days before
elective surgery4; in contrast, the Society of Thoracic Surgeons
(STS) 2005 guidelines give a class IIa recommendation to the
discontinuation of ASA 3 to 5 days before elective CABG.5
Thus, we sought to study the effect of ASA discontinuation
earlier or later than 5 days before surgery on bleeding complications, postoperative complications, and mortality in patients
undergoing isolated CABG in the current era.
Continuing medical education (CME) credit is available for this article. Go to http://cme.ahajournals.org to take the quiz.
Received March 30, 2010; accepted November 29, 2010.
From the Department of Cardiovascular Medicine (M.J., L.C.), Department of Thoracic and Cardiovascular Surgery (N.S., E.B.), and Department of
Quantitative Health Sciences (S.W.), Cleveland Clinic, Cleveland, Ohio.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.110.957373/DC1.
Correspondence to Leslie Cho, MD, Department of Cardiovascular Medicine, JB-1, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH 44195. E-mail
[email protected]
© 2011 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org
DOI: 10.1161/CIRCULATIONAHA.110.957373
577
578
Circulation
February 15, 2011
Methods
We analyzed data obtained from the Cardiovascular Information
Registry (CVIR), a database that contains clinical and laboratory
information on consecutive patients undergoing cardiothoracic surgery at Cleveland Clinic. Data from this registry have been approved
for use in research by the Institutional Review Board with patient
consent waived. This registry is certified to provide data to the STS’s
Adult Cardiac National Database. Beginning January 1, 2002,
information on preoperative use of ASA and whether ASA was
discontinued earlier than 5 days before surgery or within 5 days of
surgery was collected prospectively. From that time until January 31,
2008, 4143 patients taking preoperative ASA underwent nonemergent isolated CABG at the Cleveland Clinic. Of these, 2298 (55%)
discontinued ASA use 6 or more days before surgery (earlydiscontinuation group), and 1845 (45%) continued ASA use within 5
days of surgery (late-use group).
Clinical End Points
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
Our primary end point was a composite of in-hospital death, MI, and
stroke. Standard STS definitions of MI and stroke were used.6 MI
during the hospitalization had to fulfill 2 of the following 3 criteria:
Ischemic symptoms in the presence or absence of chest discomfort,
enzyme elevation, or at least 2 serial ECGs that showed ST-T–wave
changes from baseline.
Secondary end points were bleeding complications. These included perioperative packed red blood cell (RBC) transfusion and
return to the operating room for bleeding or tamponade. Tertiary end
points were postoperative MI, postoperative stroke, and in-hospital
mortality, individually.
Statistical Analysis
Because of substantial differences in characteristics between the 2
groups of patients, a propensity analysis was performed to reduce the
bias when outcomes were compared.7 First, multivariable logistic
regression was used to identify patient factors associated with
continued ASA use within 5 days of surgery. Variables used for this
analysis are categorized and listed in the Appendix in the online-only
Data Supplement. Of note, the variables used in our modeling were
only those collected prospectively from the CVIR database. Bagging
was used for variable selection based on the median rule8 (variables
or closely clustered variables that appeared with P⬍0.05 in 50% or
more of 500 automated stepwise bootstrap models). This parsimonious model describes factors that predict continued aspirin use.
Additional renal, coronary stenosis, and preoperative medication
variables were added to the model to create the final propensity
model. The preoperative medications included were ␤-blockers,
angiotensin-converting enzyme inhibitors, intravenous nitrates, nonwarfarin anticoagulants, warfarin, inotropes, steroids, and lipidlowering agents. In all, 31 variables were included in the propensity
model. The distribution of the propensity score in each group is
shown in Figure I in the online-only Data Supplement. A standardized difference plot (online-only Data Supplement Figure II) demonstrates that the groups were well matched with regard to variables
from the propensity model and additional demographic data.9
Using a greedy matching algorithm,10 patients from each group
were matched on similar propensity scores, and 1519 well-matched
pairs were obtained (73% of patients). These patients are shaded in
Figure I in the online-only Data Supplement, which demonstrates
that nearly all possible cases were matched, forming subsets of the
original data that were more balanced on the basis of preoperative
factors. Thereafter, primary, secondary, and tertiary outcomes were
compared to evaluate group differences.
Statistical analyses were performed with SAS version 9.1. Descriptive summary data stratified by ASA use (earlier than 5 days or
within 5 days before CABG) are presented for all variables listed in
Table 1. Continuous data are expressed as mean⫾SD. Wilcoxon rank
sum tests were used to analyze differences for the continuous data.
Categorical data are displayed as frequencies and percentages, and
comparisons were made with ␹2 tests (Fisher exact tests if appropriate). The confidence intervals used in Figure 1 were at the 68% level.
Results
Patient Demographics
Between January 1, 2002, and January 31, 2008, 4143
patients were taking ASA preoperatively and underwent
elective CABG at the Cleveland Clinic. Of those, 2298
patients (55.5%) discontinued their ASA use more than 5
days before the day of surgery (early discontinuation), and
1845 (44.5%) continued ASA use within 5 days of surgery
(late use). The demographic, clinical, and procedural data are
included in Table 1.
The early-discontinuation and late-use groups were similar
with regard to age, number of women, body mass index,
history of smoking, and history of congestive heart failure.
The early-discontinuation ASA group had higher preoperative creatinine (109⫾98.1 versus 96.4⫾61 ␮ mol/L
[1.23⫾1.11 versus 1.09⫾0.69 mg/dL], P⫽⬍0.0001) and
hematocrit levels (39.7%⫾5.32 versus 38.8%⫾5.48, P⬍0.0001).
Comorbidities differed, with the early-discontinuation group
having more patients with a history of hypertension (86%
versus 83%, P⫽0.002), treated diabetes (37% versus 33%,
P⫽0.003), peripheral arterial disease (56% versus 51%,
P⫽0.002), and renal disease (6.4% versus 3.4%, P⬍0.0001)
but fewer prior MIs (58% versus 61%, P⫽0.05). More
early-discontinuation patients had prior drug-eluting stents
placed (6% versus 2.7%, P⬍0.0001). The earlydiscontinuation patients were more likely to be taking preoperative angiotensin-converting enzyme inhibitors, lipidlowering medications, and ADP inhibitors. The late-use
group was more likely to be taking preoperative ␤-blockers
and nonwarfarin anticoagulants. Of note, the use of preoperative medications was collected as use within 24 hours of
surgery (except nonwarfarin anticoagulants and inotropes,
which were used within 48 hours).
Patients with early discontinuation were more likely to
have had prior cardiac surgery (21% versus 14%, P⬍0.0001),
with most having had 1 prior cardiac surgery. The late-use
group included more patients with at least 50% left main
coronary artery stenosis. The groups were similar with regard
to the number of coronary arteries with at least 50% stenosis
and with regard to the use of 1 or 2 internal thoracic artery
grafts. The duration of myocardial ischemia was similar, but
cardiopulmonary bypass time was slightly longer for the
early-discontinuation group (91.7⫾28.5 versus 90.2⫾27.4
minutes, P⫽0.04).
A multivariable logistic regression model was used to
determine patient factors for late ASA use (Table 2). Patients
with late ASA use were more likely to have a history of prior
MI and more severe angina (higher Canadian angina class)
and no history of hypertension. Late-use patients also were
more likely to have had a later date of surgery, no history of
prior cardiac surgery, and lower preoperative hematocrit and
creatinine values.
Results of the matching are seen in Table 3 and the Figure.
There were no statistical differences in outcomes between the
matched groups. There was no difference in the composite
Jacob et al
Table 1.
Aspirin Discontinuation Before CABG
Demographics and Procedural Characteristics of Patients Undergoing CABG
Early Discontinuation of Aspirin Use (⬎5 Days
Before Surgery)
Age, y, mean⫾SD
Late Use of Aspirin (ⱕ5 Days
Before Surgery)
Number Available
for Analysis
Value
Number Available
for Analysis
Value
P
2298
65.1⫾10.3
1845
65⫾10.8
1.00
Female, n (%)
2298
504 (22)
1845
419 (23)
0.50
Preoperative body mass index, kg/m2,
mean⫾SD)
2298
29.7⫾5.7
1845
29.4⫾5.48
0.09
Preoperative creatinine, ␮mol/L
(mg/dL), mean⫾SD
2282
109⫾98.1 (1.23⫾1.11)
1832
96.4⫾61 (1.09⫾0.69)
⬍0.0001
39.7⫾5.32
1818
38.8⫾5.48
⬍0.0001
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
Preoperative hematocrit, mean⫾SD (%)
2255
History of MI, n (%)
2298
1325 (58)
1845
1120 (61)
0.05
History of CVA, n (%)
2298
231 (10)
1845
154 (8.3)
0.06
History of hypertension, n (%)
2295
1978 (86)
1844
1523 (83)
0.002
History of treated diabetes, n (%)
2296
855 (37)
1845
605 (33)
0.003
History of PVD, n (%)
2298
1290 (56)
1845
946 (51)
0.002
History of CHF, n (%)
2298
552 (24)
1845
414 (22)
0.20
History of smoking, n (%)
2295
1462 (64)
1844
1168 (63)
History of renal disease,* n (%)
2298
147 (6.4)
1845
63 (3.4)
History of percutaneous intervention,
n (%)
2298
665 (29)
1845
492 (27)
PTCA, n (%)
583 (25)
Drug-eluting stent,† n (%)
137 (6)
440 (24)
49 (2.7)
0.80
⬍0.0001
0.10
0.30
⬍0.0001
Bare-metal stent,† n (%)
109 (4.7)
67 (3.6)
0.08
Stent (unknown type), n (%)
315 (14)
265 (14)
0.60
262 (14)
⬍0.0001
⬍0.0001
History of cardiac surgery, n (%)
2298
No. of prior cardiac operations, n (%)
2298
478 (21)
1845
1845
0
1820 (79)
1583 (86)
1
401 (17)
231 (13)
2
70 (3)
3
No. of coronary arteries with ⱖ50%
stenosis, n (%)
29 (1.6)
7 (0.3)
2272
2 (0.11)
1828
0.90
0
27 (1.2)
20 (1.1)
1
171 (7.5)
129 (7.1)
2
483 (21)
394 (22)
3
1591 (70)
1285 (70)
Left main stenosis ⱖ50%, n (%)
2247
611 (27)
1810
571 (32)
0.002
No. of internal thoracic grafts, n (%)
2298
205 (8.9)
1845
126 (6.8)
0.007
1
1737 (76)
1384 (75)
2
356 (15)
335 (18)
␤-blocker use, n (%)
2279
1842 (81)
1823
1544 (85)
0.001
ACE inhibitor use, n (%)
2279
1442 (63)
1823
1097 (60)
0.04
Lipid-lowering medication use, n (%)
2279
1022 (45)
1823
488 (27)
⬍0.0001
ADP inhibitor use, n (%)
2279
569 (25)
1823
265 (15)
⬍0.0001
Anticoagulant use (excluding
warfarin), n (%)
2279
688 (30)
1823
898 (49)
⬍0.0001
Warfarin use, n (%)
2279
65 (2.9)
1823
39 (2.1)
0.10
CVA indicates cerebrovascular accident; PVD, peripheral vascular disease; CHF, congestive heart failure; PTCA, percutaneous transluminal coronary
angioplasty; and ACE, angiotensin-converting enzyme.
Medication data refer to use within 24 hours of surgery, except for anticoagulants, which refers to use within 48 hours. Warfarin, specifically, refers
to use within 24 hours.
*History of renal disease defined as documented history of renal failure and/or a history of creatinine ⬎2.0 or on dialysis.
†Data on whether patient had either a drug-eluting or a bare-metal stent was collected beginning in 2004.
579
580
Circulation
February 15, 2011
Table 2. Factors Associated With Late Aspirin Use (<5 Days)
and CABG
Factor
Coefficient⫾SE
History of MI
P
R (%)*
0.20⫾0.068
0.003
89
0.083⫾0.029
0.005
80
Date of surgery (earlier)
⫺0.24⫾0.019
⬍0.0001
95
History of cardiac surgery
⫺0.57⫾0.089
⬍0.0001
81
History of hypertension
⫺0.22⫾0.089
0.01
61
History of COPD
⫺0.24⫾0.086
0.005
57
Preoperative creatinine (lower)
⫺0.13⫾0.047
0.007
50
Preoperative hematocrit (lower)
⫺0.027⫾0.0064
⬍0.0001
96
Preoperative BUN (lower)
⫺0.012⫾0.0040
0.004
64
Canadian angina class (higher)
SE indicates standard error; COPD, chronic obstructive pulmonary disease;
and BUN, blood urea nitrogen.
*Reliability is percentage of 1000 bootstrap models that the variable entered.
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
end point, with 26 events (1.7%) in the early-discontinuation
group versus 28 (1.8%) in the late-use group (P⫽0.80). More
intraoperative RBC transfusions (23% versus 20%, P⫽0.03)
and postoperative transfusions (30% versus 26%, P⫽0.009)
were used in the late-use group (Table 3). The rate of
reoperation for bleeding was 2.4% versus 3.4% in the
early-discontinuation and late-use groups, respectively
(P⫽0.10). There was no difference between early discontinuation and late use with regard to the individual outcomes of
postoperative cerebrovascular accident (0.92% versus 0.79%,
P⫽0.70), postoperative MI (0.33% versus 0.39%, P⫽0.80),
or in-hospital death (0.66% versus 0.72%, P⫽0.80; Figure).
Postoperative length of stay was similar; the median length of
stay was 6 days in both groups (P⫽0.50).
Table 3.
Figure. Adjusted analysis of ASA discontinuation and postCABG outcomes. Outcomes shown with 68% confidence intervals. CVA indicates cerebrovascular accident; OR, operating
room. Composite outcome includes in-hospital death, postoperative MI, and postoperative CVA.
Discussion
The guidelines regarding the timing of ASA discontinuation
before CABG or whether it should be discontinued at all are
varied. The American College of Cardiology/American Heart
Association guidelines advocate discontinuation 7 to 10 days
before surgery, whereas the STS advocates discontinuation 5
days before elective CABG.4 This is driven by concern about
increased postoperative bleeding complications. Both guidelines quote a study by Kallis et al11 in 1994 in which 100
patients undergoing elective CABG randomly received 300
mg of ASA or placebo for 2 weeks before surgery. They
found that patients taking ASA had increased blood loss
Transfusion Use Based on Timing of Aspirin Discontinuation, Matched Groups
Early Discontinuation of
Aspirin Use (⬎5 Days
Before Surgery), n
Patients who received an
intraoperative RBC transfusion
1518
No. of intraoperative RBC
units used
1518
0
Count (%)
Late Use of Aspirin
(ⱕ5 Days Before
Surgery), n
Count (%)
P
301 (20)
1516
351 (23)
0.03
1516
1217 (80)
0.10
1165 (77)
1
83 (5.5)
86 (5.7)
2
124 (8.2)
163 (11)
3
45 (3)
45 (3)
4
27 (1.8)
37 (2.4)
ⱖ5
22 (1.4)
Patients who received a
postoperative RBC transfusion
1518
No. of postoperative RBC
units used
1518
0
390 (26)
20 (1.3)
1516
454 (30)
1516
0.009
0.02
1128 (74)
1062 (70)
1
164 (11)
166 (11)
2
130 (8.6)
165 (11)
3
37 (2.4)
45 (3)
4
15 (0.99)
32 (2.1)
ⱖ5
44 (2.9)
46 (3)
Jacob et al
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
(measured from mediastinal or chest tubes), received more
RBC transfusions, and had more reoperations for bleeding. In
addition, the STS found 4 more randomized controlled trials
that suggested that preoperative use of ASA resulted in
increased bleeding complications. Except for the study by
Kallis et al, ASA was started within 48 hours of surgery,
sometimes as late as the night before.11–14
Both the American College of Cardiology/American Heart
Association and the STS base the timing of ASA discontinuation on platelet physiology, because ASA is an irreversible
cyclooxygenase inhibitor. New platelets need to be formed to
overcome the effect of ASA. The lifespan of a platelet is
approximately 7 to 10 days,15 which influenced the American
College of Cardiology/American Heart Association recommendation. However, it takes 3 to 5 days for half of the
platelet pool to be replenished, which may be enough to
normalize bleeding time and thromboxane B2 levels, and
thus, the STS recommends 5 days as the cutoff.5
In the present study of 4143 patients undergoing CABG,
we found there was no significant difference with regard to
timing of ASA discontinuation in the composite primary end
point or the individual postoperative complications of MI,
cerebrovascular accident, or in-hospital death. Late use was
associated with increased use of RBC transfusions. Although
there was no statistically significant difference in reoperation
rate, there was a trend toward more reoperations in the
late-use group that may correlate with increased use of RBC
transfusion in this group. Ideally, a larger population would
be used to study this effect. The patient group in the present
study represents the largest group of patients studied with
regard to bleeding and postoperative outcomes after CABG
from a single center. There were no patients in either group
who had reoperation for graft occlusion. Because of the
significant difference in patient characteristics between
groups, propensity analysis was performed, and patients were
matched on the basis of their propensity score. The maximal
number of patients was matched for each score, including
those at both extremes. As seen in Figure II in the online-only
Data Supplement, the groups were well matched with regard
to standardized differences in certain variables. The distribution of scores was similar in both groups around a score of 50,
without skew in either direction, which implies that there was
equipoise with regard to timing of ASA discontinuation. To
the best of our knowledge, this is the largest study to date to
address the issue of early or late ASA use before CABG.
Many studies in recent years have advocated the restricted
use of intraoperative and postoperative blood transfusion and
the conservation of blood intraoperatively. This restricted use
of RBC transfusions in particular is guided by concern about
a possible increase in postoperative infection (bacteremia and
superficial and deep sternal wound infections),16 acute lung
injury,17 and all-cause mortality. In particular, Koch et al18
found that transfusion of RBCs was associated with increased
mortality at 6 months, 5 years, and 10 years.
Some may argue that an increased rate of perioperative
transfusion would support early discontinuation. Given the
lack of difference in the incidence of reoperation for bleeding,
we would support the late use of ASA in high-risk patients to
reduce the rate of preoperative cardiovascular events. With
Aspirin Discontinuation Before CABG
581
our database, we were unable to account for those patients
who were scheduled for CABG but may have had surgery
cancelled or delayed owing to cardiac events. There is a
possible risk of preoperative MI or death if ASA use is
stopped before CABG is performed.
Two systematic reviews have observed the consequences
of ASA withdrawal. One by Burger et al19 reported 3
retrospective studies that included patients undergoing
chronic ASA therapy who presented with acute MI, acute
coronary syndrome, and acute lower-limb ischemia. Among
these patients, ASA withdrawal was associated with a 2.3%
to 10.2% incidence of an acute coronary event and a 6.1%
incidence of acute lower-limb ischemia. The mean time from
ASA withdrawal to acute coronary syndrome was 8.5⫾3.6
days. Most of these patients (53.9%) discontinued ASA for
upcoming noncardiac surgery. The other systematic review
included 6 studies and 50 279 patients taking ASA for
secondary prevention of coronary artery disease, acute coronary syndrome, CABG, and placement of a drug-eluting
stent.20 There was an increased risk of major adverse cardiovascular events with ASA discontinuation (odds ratio 3.14,
95% confidence interval 1.75 to 5.61, P⫽0.0001). Specifically, for patients taking ASA for secondary prevention of
acute coronary syndrome, the risk for major adverse cardiovascular events was increased (odds ratio 1.82, 95% confidence interval 1.52 to 2.18, P⬍0.00001). In the era of
drug-eluting stents, with their increased incidence of late stent
thrombosis, the issue of late ASA use has become more
pertinent. Recent data suggest that patients are at highest risk
of stent thrombosis if both antiplatelet agents (ASA and
clopidogrel) are discontinued.21 Thus, keeping at least 1
antiplatelet agent “on board” may reduce this complication
before surgery.
Currently, a randomized controlled trial is enrolling patients undergoing isolated, nonurgent CABG (on-pump and
off-pump) to observe the effect of aspirin use (given just
before surgery) and the operative use of tranexamic acid on
all-cause mortality, ischemic complications, and bleeding
outcomes at 30 days after surgery.22 The Aspirin and Tranexamic Acid for Coronary Artery Surgery (ATACAS) Trial
plans to include 4600 patients from several centers in Australia, New Zealand, Asia, and Europe and has already
enrolled approximately 900 patients.22,23 It is hoped that
ATACAS will yield data on the use of antiplatelet and
antifibrinolytic agents immediately before surgery; however,
it will not answer the question about the preoperative use of
aspirin, because the late use of aspirin (within 5 days of
surgery) is an exclusion criterion.
Study Limitations
One major limitation of the CVIR database is that it includes
patients who have had surgery, but it does not include those
who had CABG planned but did not undergo surgery owing
to a preoperative illness that made surgery too much of a risk.
If all patients who had planned CABG were included in the
cohort, we might have found that early ASA discontinuation
was associated with preoperative MI or stroke. Also, this was
a single-center study. The CVIR database did not collect
582
Circulation
February 15, 2011
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
information on ASA dose at the time of discontinuation or on
exact ADP inhibitor use. The timing of ASA discontinuation
was left to the discretion of individual surgeons. In 2008, we
began to collect data regarding the dose of aspirin used as
well as the actual day it was discontinued relative to the
surgical date. These data were not available for the present
study population (January 1, 2002, to January 31, 2008). We
were unable to adjust for clopidogrel use; ideally, we would
collect data on clopidogrel use within 5 days of surgery. Also,
we do not have data on the effect of ASA dose on bleeding
complications; presumably, there is no difference. We are
currently collecting these data for patients undergoing openheart surgery at our institution. In addition, the CVIR now
collects data on medications used within 30 days before
surgery and records the date of cessation as well. Antifibrinolytic use is not captured in the CVIR database. Except for an
early study in the 1980s, aprotinin has not been used regularly
at the Cleveland Clinic Foundation. Aminocaproic acid is
used routinely for all CABG cases at the Cleveland Clinic
Foundation as a loading dose, then given as an infusion
during the case.
Past studies that argued for early discontinuation focused
on several markers of increased bleeding. Here, we were able
to report the number of transfusions used and the rate of
reoperation for bleeding. We did not have data for the total
amount of chest tube drainage or the rate of chest tube
drainage. It is presumed that very high chest tube drainage or
fast drainage would prompt reexploration in certain cases,
transfusion of blood products, or both.
There are inherent limitations on the generalizability of a
retrospective study regarding the discontinuation of ASA
before CABG. A randomized, controlled trial would be ideal
to control for unforeseen confounders. With the use of a
propensity model to match the populations, we tried to come
as close as possible to comparing groups similar in all ways
except for their timing of ASA discontinuation.
Conclusions
The late use of ASA is associated with an increased use of
transfusions but no statistically significant difference in
reoperations for bleeding. There was no difference in postoperative MI or stroke or in-hospital all-cause mortality.
Thus, on the basis of the present study, we recommend that
clinicians weigh the risks and benefits of late ASA use on the
basis of the patient’s risk profile before CABG.
Acknowledgment
We would like to thank Kathryn Brock for her review of this
manuscript.
Disclosures
None.
References
1. Mangano DT; Multicenter Study of Perioperative Ischemia Research
Group. Aspirin and mortality from coronary bypass surgery. N Engl
J Med. 2002;347:1309 –1317.
2. Dacey LJ, Munoz JJ, Johnson ER, Leavitt BJ, Maloney CT, Morton JR,
Olmstead EM, Birkmeyer JD, O’Connor GT. Effect of preoperative
aspirin use on mortality in coronary artery bypass grafting patients. Ann
Thorac Surg. 2000;70:1986 –1990.
3. Bybee KA, Powell BD, Valeti U, Rosales AG, Kopecky SL, Mullany C,
Wright RS. Preoperative aspirin therapy is associated with improved
postoperative outcomes in patients undergoing coronary artery bypass
grafting. Circulation. 2005;112(suppl):I-286 –I-292.
4. Eagle KA, Guyton RA, Davidoff R, Edwards FH, Ewy GA, Gardner TJ,
Hart JC, Herrmann HC, Hillis LD, Hutter AM Jr, Lytle BW, Marlow RA,
Nugent WC, Orszulak TA, Antman EM, Smith SC Jr, Alpert JS,
Anderson JL, Faxon DP, Fuster V, Gibbons RJ, Gregoratos G, Halperin
JL, Hiratzka LF, Hunt SA, Jacobs AK, Ornato JP. ACC/AHA 2004
guideline update for coronary artery bypass graft surgery: summary
article: a report of the American College of Cardiology/American Heart
Association Task Force on Practice Guidelines (Committee to Update the
1999 Guidelines for Coronary Artery Bypass Graft Surgery). J Am Coll
Cardiol. 2004;44:e213– e310.
5. Ferraris V, Ferraris S, Moliterno D, Camp P, Walenga J, Messmore H,
Jeske W, Edwards F, Royston D, Shahian D. The Society of Thoracic
Surgeons practice guideline series: aspirin and other antiplatelet agents
during operative coronary revascularization (executive summary). Ann
Thorac Surg. 2005;79:1454 –1461.
6. STS General Thoracic Data Specifications, version 2.081. September
5, 2008. Available at: http://www.sts.org/documents/pdf/ndb/
GeneralThoracicDataSpecsV2081.pdf. Accessed February 24, 2010.
7. Blackstone EH. Comparing apples and oranges. J Thorac Cardiovasc
Surg. 2002;123:8 –15.
8. Breiman L. Bagging predictors. Mach Learn. 1996;24:123–140.
9. Austin PC, Mamdani MM. A comparison of propensity score methods: a
case-study estimating the effectiveness of post-AMI statin use. Stat Med.
2006;25:2084 –2106.
10. Bergstralh EJ, Konsanke JL. Technical Report Series No. 56, Computerized Matching of Cases to Controls. Rochester, MN: Department of
Health Science Research, Mayo Clinic; 1995.
11. Kallis P, Tooze J, Talbot S, Cowans D, Bevan D, Treasure T. Preoperative aspirin decreases platelet aggregation and increases postoperative blood loss: a prospective, randomised, placebo controlled,
double-blind clinical trial in 100 patients with chronic stable angina. Eur
J Cardiothorac Surg. 1994;8:404 – 409.
12. Ferraris SP, Lough FC, Berry WR, Ferris VA. Preoperative aspirin
ingestion increases operative blood loss after coronary artery bypass
grafting. Ann Thorac Surg. 1988;45:71–74.
13. Goldman S, Copeland J, Mortiz T, Henderson W, Zadina K, Ovitt T, Doherty
J, Read R, Chesler E, Sako Y. Improvement in early saphenous vein graft
patency after coronary artery bypass surgery with antiplatelet therapy: results
of a Veterans Administration Cooperative Study. Circulation. 1988;77:
1324–1332.
14. Sethi GK, Copeland JG, Goldman S, Mortiz T, Zadina K, Henderson
WG. Implications of preoperative administration of aspirin in patients
undergoing coronary artery bypass grafting: Department of Veterans
Affairs Cooperative Study on Antiplatelet Therapy. J Am Coll Cardiol.
1990;15:15–20.
15. McMillan R. Hemorrhagic disorders: abnormalities of platelet and
vascular function. In: Goldman L, Ausiello D, eds. Cecil Medicine. 23rd
ed. Philadelphia, PA: Saunders Elsevier; 2007.
16. Banbury MK, Brizzio ME, Rajeswaran J, Lytle BW, Blackstone EH.
Transfusion increases the risk of postoperative infection after cardiovascular surgery. J Am Coll Surg. 2006;202:131–138.
17. Gardner TJ. To transfuse or not to transfuse. Circulation. 2007;116:
458 – 460.
18. Koch C, Li L, Duncan A, Mihaljevic T, Loop F, Starr N, Blackstone E.
Transfusion in coronary artery bypass grafting is associated with reduced
long-term survival. Ann Thorac Surg. 2006;81:1650 –1657.
19. Burger W, Chemnitius JM, Kneissl GD, Rucker G. Low-dose aspirin for
secondary cardiovascular prevention: cardiovascular risks after its perioperative withdrawal versus bleeding risks with its continuation: review
and meta-analysis. J Intern Med. 2005;257:399 – 414.
20. Biondi-Zoccai GG, Lotrionte M, Agostoni P, Abbate A, Fusaro M,
Burzotta F, Testa L, Sheiban I, Sangiorgi G. A systematic review and
meta-analysis on the hazards of discontinuing or not adhering to aspirin
among 50,279 patients at risk for coronary artery disease. Eur Heart J.
2006;27:2667–2674.
Jacob et al
21. Eisenberg MJ, Richard PR, Libersan D, Filion KB. Safety of short-term
discontinuation of antiplatelet therapy in patients with drug-eluting stents.
Circulation. 2009;119:1634 –1642.
22. Myles PS, Smith J, Knight J, Cooper DJ, Silbert B, McNeil J, Esmore DS,
Buxton B, Krum H, Forbes A, Tonkin A. Aspirin and Tranexamic Acid
Aspirin Discontinuation Before CABG
583
for Coronary Artery Surgery (ATACAS) Trial: rationale and design. Am
Heart J. 2008;155:224 –230.
23. ATACAS: Aspirin and Tranexamic Acid for Coronary Artery Surgery: a
randomised controlled trial. Available at: http://www.atacas.org.au/
centres. Accessed June 26, 2010.
CLINICAL PERSPECTIVE
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
The use of aspirin for patients with proven coronary artery disease is nearly ubiquitous, especially in those undergoing
revascularization, whether percutaneous or surgical. The American Heart Association (AHA), American College of
Cardiology (ACC), and Society of Thoracic Surgeons (STS) have given guidance as to the use of aspirin prior to coronary
artery bypass grafting based on evidence mostly collected in the 1980s and 1990s. These guidelines are influenced by
concerns of increased bleeding in the postoperative period and differ between societies. Thus, in the elective coronary
artery bypass grafting population, aspirin is routinely discontinued up to 1 week prior to surgery. However, there is
increasing concern that the discontinuation of aspirin, especially in patients with prior percutaneous coronary intervention,
is linked to increased myocardial infarction, stroke, and death. More recently, there have been studies suggesting increased
mortality in those who discontinue aspirin early before surgery. In our study of ⬎4000 patients undergoing elective,
isolated coronary artery bypass grafting, there was no significant difference between those with early discontinuation of
aspirin (ⱖ6 days before surgery) and late aspirin use (within 5 days) with regards to the composite outcome of in-hospital
mortality, myocardial infarction, and stroke. Late use was associated with more intraoperative transfusion and
postoperative transfusion but similar number of reoperations for bleeding. Thus, late use of aspirin results in no difference
in the postoperative cardiovascular outcomes; however, there is an increased risk of bleeding and transfusion requirements.
Go to http://cme.ahajournals.org to take the CME quiz for this article.
Effect of Timing of Chronic Preoperative Aspirin Discontinuation on Morbidity and
Mortality in Coronary Artery Bypass Surgery
Miriam Jacob, Nicholas Smedira, Eugene Blackstone, Sarah Williams and Leslie Cho
Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2017
Circulation. 2011;123:577-583; originally published online January 31, 2011;
doi: 10.1161/CIRCULATIONAHA.110.957373
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2011 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/123/6/577
Data Supplement (unedited) at:
http://circ.ahajournals.org/content/suppl/2011/01/27/CIRCULATIONAHA.110.957373.DC1
http://circ.ahajournals.org/content/suppl/2013/10/17/CIRCULATIONAHA.110.957373.DC2
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/
Effect of Timing of Chronic Preoperative Aspirin Discontinuation on Morbidity and
Mortality in Coronary Artery Bypass Surgery
SUPPLEMENTAL MATERIAL
Miriam Jacob, MD, Nicholas Smedira MD, Eugene Blackstone MD,
Sarah Williams MS, Leslie Cho, MD
Cleveland Clinic
From the Department of Cardiovascular Medicine (M.J, L.C.), Department of Thoracic
and Cardiovascular Surgery (N.S., E.B.) and Department of Quantitative Health Sciences
(S.W.)
Cleveland, Ohio
Corresponding Author
Leslie Cho MD
Department of Cardiovascular Medicine, JB-1
Cleveland Clinic
9500 Euclid Ave
Cleveland, OH 44195
216-445-6320
216-444-8856 fax
[email protected]
eFigure Legend:
eFigure 1. Mirrored histogram of distribution of propensity scores for Aspirin
Usage groups.
Mirrored histogram of distribution of propensity scores for Early Discontinuation
(bars above zero line) and Late Use (bars below zero line). The darkened area
represents matched patient pairs, showing that they cover the complete spectrum
of cases.
eFigure 2. Covariate balance description before and after matching between Early
Discontinuation versus Late Use groups.
Values in the x-axis represent the percent standardized difference between Early
Discontinuation and Late Use groups.
eFigure 1
eFigure 2
Appendix 1: Variables considered for multivariable logistic regression
An asterisk denotes variables used in the final propensity model.
Demographic:
Age (years)*, gender*, weight (kg), height (cm), body
surface area (m2), body mass index (kg•m-2)*.
Symptoms:
NYHA functional class (I-IV), Canadian Angina Class*.
Ventricular dysfunction:
degree of left ventricular dysfunction (1=none, 2=mild,
3=moderate, 4=severe)*, Ejection fraction (%),
Hypertension*.
Valve Pathology:
Aortic valve: regurgitation*, stenosis,
Mitral valve: regurgitation*, stenosis,
Tricuspid valve: regurgitation.
Coronary anatomy:
Left main trunk disease (maximum % stenosis)*, left
anterior descending coronary artery system disease
(maximum % stenosis)*, right coronary artery system
disease (maximum % stenosis)*, left circumflex coronary
artery system disease (maximum % stenosis)*.
Other cardiac comorbidity: Atrial fibrillation, history of MI*, history of cardiac
disease, history of cardiac surgery*, complete heart
block/Pacer*, ventricular arrhythmia.
Noncardiac comorbidity:
History of diabetes*, history of peripheral vascular
disease*, history of smoking, carotid disease, COPD*,
popliteal disease, peripheral vascular disease, cholesterol
values, triglycerides, BUN*, creatinine*, bilirubin*,
hematocrit*.
Experience:
Date of operation (years since 1/1/1997)*.
Medication Use:
Beta Blocker*, ACE inhibitor*, IV Nitrate*,
Anticoagulant*, Warfarin*, Inotrope*, Steroid*, Lipid
Lowering Medication*.
Page 444
Résumés d’articles
Influence du délai d’interruption préopératoire du traitement chronique
par l’aspirine sur la morbidité et la mortalité chez les patients
relevant d’un pontage aorto-coronaire
Miriam Jacob, MD ; Nicholas Smedira, MD ; Eugene Blackstone, MD ; Sarah Williams, MS ; Leslie Cho, MD
Contexte—Il a été établi que l’administration d’acide acétylsalicylique (AAS) diminue la mortalité et l’incidence des événements ischémiques
après réalisation d’un pontage aorto-coronaire (PAC) ; les avis divergent toutefois quant à l’intervalle qu’il y a lieu de respecter
entre l’interruption du traitement chronique par l’AAS et l’intervention chirurgicale pour prévenir les complications hémorragiques
postopératoires. Nous avons donc utilisé la base de données du registre des informations cardiovasculaires de la Cleveland Clinic (Ohio,
Etats-Unis) pour évaluer l’influence exercée par le délai d’arrêt de l’AAS avant la réalisation du PAC sur l’incidence des événements
cardiovasculaires majeurs et des hémorragies postopératoires.
Méthodes et résultats—Entre le 1er janvier 2002 et le 31 janvier 2008, 4 143 patients ont fait l’objet d’un PAC à la Cleveland Clinic alors qu’ils
avaient pris de l’AAS au long cours avant l’intervention. Parmi eux, 2 298 avaient arrêté l’AAS au moins 6 jours avant d’être opérés
(interruption précoce) et les 1 845 autres avaient continué à prendre le médicament au cours des 5 jours ayant précédé l’intervention
(poursuite tardive). En raison des importantes différences existant entre ces deux groupes, nous avons procédé à une analyse du score de
propension et à un appariement en fonction de 31 variables afin de pouvoir comparer les taux d’événements de façon équitable. Nous avons
ainsi constitué 1 519 paires de patients parfaitement concordantes (représentant 73 % de la cohorte initiale). Aucune différence significative
n’a été relevée entre les patients qui avaient interrompu précocement leur traitement par l’AAS et ceux qui l’avaient poursuivi tardivement en
termes d’incidence du critère de jugement composite regroupant le décès pendant l’hospitalisation, l’infarctus du myocarde et l’accident
vasculaire cérébral (1,7 % versus 1,8 % ; p = 0,80). La poursuite tardive du traitement a augmenté les taux de recours aux transfusions
peropératoires (23 % versus 20 % ; p = 0,03) et postopératoires (30 % versus 26 % ; p = 0,009), mais le nombre de réinterventions motivées
par des saignements a été comparable dans les deux groupes (3,4 % contre 2,4 % ; p = 0,10).
Conclusions—Chez des patients ayant fait l’objet d’un PAC isolé, l’arrêt tardif de l’AAS n’a pas eu d’influence sur l’incidence des événements
cardiovasculaires postopératoires ; en revanche, les besoins en transfusions ont été plus importants. C’est pourquoi, chez de tels sujets,
nous recommandons de mettre en balance les risques et les bénéfices de la poursuite tardive du traitement par l’AAS. (Traduit de l’anglais :
Effect of Timing of Chronic Preoperative Aspirin Discontinuation on Morbidity and Mortality in Coronary Artery Bypass Surgery. Circulation.
2011;123:577–583.)
Mots clés : aspirine 䊏 artériosclérose 䊏 pontage 䊏 maladie coronaire 䊏 chirurgie
Le gène Notch1 exprimé par les cellules médullaires commande la
réparation cardiaque après infarctus du myocarde
Yuxin Li, MD, PhD ; Yukio Hiroi, MD, PhD ; Soeun Ngoy, BS ; Ryuji Okamoto, MD, PhD ;
Kensuke Noma, MD, PhD ; Chao-Yung Wang, MD ; Hong-Wei Wang, MD, PhD ;
Qian Zhou, MD ; Freddy Radtke, PhD ; Ronglih Liao, PhD ; James K. Liao, MD
Contexte—On connaît mal les voies de signalisation qui régulent le recrutement des cellules médullaires (CM) vers un territoire myocardique
lésé. Les récepteurs Notch régissent la détermination binaire du destin cellulaire et pourraient également moduler la fonction des CM. On
ignore toutefois si, par les signaux qu’il envoie à ces cellules, le gène Notch1 joue un rôle de commande de la régénération des tissus
myocardiques lésés.
Méthodes et résultats—Les souris chez lesquelles une délétion cardiosélective de Notch1 a été pratiquée après leur naissance se montrent
semblables aux animaux témoins en termes de taille d’infarctus et de fonction cardiaque après induction d’une lésion ischémique. En
revanche, les souris porteuses d’une délétion hémizygote globale de Notch1 (N1± ) présentent des infarctus plus étendus et une fonction
cardiaque plus fortement altérée. Après que la moelle osseuse (MO) de souris N1± a été transplantée chez des souris de type sauvage (TS), les
infarctus développés par ces dernières ont été de plus grande taille, leur fonction cardiaque a présenté une dégradation plus marquée et la
formation de néovaisseaux dans la zone bordant l’infarctus a été diminuée comparativement à ce qui a été observé chez des souris TS ayant
reçu de la MO provenant d’homologues TS. En revanche, la greffe de MO de souris TS chez des animaux N1± a atténué les lésions
myocardiques engendrées chez ces derniers. De fait, la délétion hémizygote du gène Notch1 au sein des CM contribue à diminuer le
recrutement, la prolifération et la survie des cellules souches mésenchymateuses (CSM). Comparativement à celle de CSM de souris TS,
l’injection de CSM provenant d’animaux N1± dans le territoire myocardique infarci a pour effet de majorer la lésion tissulaire, alors que
l’injection de CSM surexprimant le domaine intracellulaire de Notch1 diminue la taille d’infarctus et améliore la fonction cardiaque.
Conclusions—Ces observations montrent que les signaux émis par Notch1 en direction des CM jouent un rôle majeur dans la réparation
cardiaque et que, dès lors, l’approche consistant à intensifier ces signaux pourrait présenter un intérêt thérapeutique chez les patients atteints
d’une affection cardiaque d’origine ischémique. (Traduit de l’anglais : Notch1 in Bone Marrow–Derived Cells Mediates Cardiac Repair After
Myocardial Infarction. Circulation. 2011;123:866–876.)
Mots clés : cellules souches 䊏 thérapie génique 䊏 infarctus du myocarde 䊏 angiogenèse
© 2011 Lippincott Williams & Wilkins
Circulation est disponible sur http://circ.ahajournals.org
444
12:43:12:10:11
Page 444